D ATA PROCESSING AND RELIABILITY 1 D ATA ANALYSIS PROCEDURES

There are various ways of approaching qualitative data analysis, taking into account the narrative approach, analysis needed to reflect the deep meaning generated in the research process. (Atkinson & Delamont, 2005). To do this Titchener's (2010) framework of: data immersion, data coding, and data reduction was loosely followed, influenced by thematic analysis and concept mapping. The first step in data immersion is the transcribing of recorded data. In this study all but one of the interviews and focus groups were voice recorded, and while the transcription of these took time it was key in becoming 'familiar' with the data. Following this transcription a process of reading and re-reading is recommended for immersion and allows the researcher to

think over and reflect on their initial insights from the data (Hunter, Lusardi, Zucker, Jacelon, & Chandler, 2002; Willis, 2010). The transcription and immersion process provided insights into some commonalities in what participants shared and possible ways of structuring the findings, giving a good foundation to move onto data coding.

Data coding and reduction were influenced by thematic analysis in which, "data is analysed by themes through an inductive process" (Titchener, 2010, p. 37). It was inductive in the sense that the meaning and patterns in the data drive the themes, rather than finding themes to test a theory or fit predetermined categories (Natalier, 2010). That said, themes do need to, "capture something important about the data in relation to the research question, and represent some level of patterned response or meaning within the data set", in this sense the research questions were used to guide the categorising of themes (Braun & Clarke, 2006). Data coding, which is basically a categorising stage in the analysis, was done first, by going through each transcript and writing various summary words or phrases beside sentences and paragraphs. Following this summary words and phrases were used to come up with some initial codes (Willis, 2010)

Data reduction was the final step and involved "exploring relationships" between codes in order to reduce the data into manageable and meaningful themes (Willis, 2010, p. 421). It is during this stage that much referral back to the 'bigger picture' in terms of research aims and questions was done (Titchener, 2010). Conceptual mapping was used as the data reduction technique in this study and allowed for a visual representation of the various ideas that emerged in the research and exploration into how these were connected. Codes were then grouped into themes and sub-themes. I agree with Braun and Clarke (2006) who argue that thematic analysis involves the defining and redefining of themes in a cyclical manner until the researcher is happy that the themes do justice to the knowledge shared. Doing justice to what participants shared was important throughout the process. A further concern was to avoid being reductionist and flattening the findings, in light of the depth of meaning evident in the data and the complexity, interconnectedness and 'energy' the information held for the participants. For this reason constant referral back to the transcripts was made to check that meaning had been captured correctly. Additionally, direct quotes from participants were used throughout the results section to allow their voice to come through.

4.4.2R

Qualitative research differs from quantitative research in terms of assessing validity and reliability of findings, but this is no less (O'Leary, 2009). Producing dependable findings was of concern for this study to ensure that the findings were trustworthy and honoured to the participants narratives. Research can be described as dependable and trustworthy, "if the reader is able to audit the events, influences and actions of the researcher" (Koch, 2006, p. 91). Various criteria for judging trustworthiness of qualitative research findings have been forthcoming. For the purposes of this thesis, an adaption of O'Learly's (2009) framework for credibility was used and involved three factors: managing subjectivities; methodological consistency and verifiability; and authenticity of data.

It has already been mentioned that acknowledging personal subjectivities is important within qualitative research with regards to ensuring integrity of the research process (O'Leary, 2009). This transparency in acknowledging subjectivities, personal positionality and potential biases was the first step in conducting dependable and ethical research. Secondly, detailing how personal subjectivities are mediated for, and ethical research conducted in light of them, is another important step. This was undertaken in the current research by discussing my positionality within the research and detailing, through the narrative inquiry methodological framework, the principles and procedures used to manage potential biases.

In terms of methodological consistency and verifiability, research should never be a "haphazard activity" (O'Leary, 2009, p. 59). A well-designed and well-documented research process reinforces the dependability of findings (Koch, 2006). A lot of thought, as well as consultation with Solomon Islanders, went into the design of the current study. This was important in ensuring ethical and systematic research procedures could be followed that remained true to the research questions and the theoretical and philosophical background of the study. This design was then clearly documented in this chapter through outlining the methodological framework and approach. In addition, how the research played out in practice, including challenges encountered, and adaption of research techniques, were also outlined.

Regarding authenticity of the findings presented, qualitative research is not as concerned with generalisability of findings as quantitative research, nor with measuring how well 'truth' about the phenomena being studied has been captured (Walter, 2010b). Rather, acknowledgement is made from the beginning of people's unique

knowledge and negotiation of meaning, as well as, of the fact that narratives created during research are "situated interactive performances" (Chase, 2005, p. 657; Sapsford & Jupp, 1996). The "truth value" and authenticity of qualitative research, therefore, comes from ensuring that findings presented are true to the 'data' collected and that a range of voices and techniques are drawn upon in the research to gather this data (O'Leary, 2009, p. 61). The concept of 'triangulation', "using more than one source of data to confirm the authenticity of each source", is important here (O'Leary, 2009, p. 115). For this study triangulation occurred through the use of both one-on-one and group stori sessions, and though consulting with a range of 'voices', including: youth, community members, and other key stakeholders who work with youth. Authenticity is also linked to presenting findings that do justice to the 'data' collected (Koch, 2006). The transcripts allowed constant referral back to the 'raw data' (O'Leary, 2009). Additionally, sending transcripts and results chapters to participants allowed them to verify the authenticity of the findings.

4.5CHAPTER SUMMARY

In summary, this chapter outlined the methodological approach undertaken in this study, including fieldwork experiences and data analysis techniques, thereby setting the scene for the presentation of findings. The constructivist/interpretivist philosophical standpoint, which holds that people are continually constructing meaning and acting and interacting based on their interpretations of the world, and the research aims, objectives and questions, all guided the methodological approach undertaken for this study. The narrative inquiry methodological framework, influenced by principles of indigenous Pacific methodologies and actor-oriented approaches, together with the ethical considerations made in the study, provided the foundation from which to enter the 'field'. Through obtaining the Solomon Islands Research Permit, and the access to participants through the Ola Fou network and government departments, I was able to

stori with a wide range of people around the issue of youth agency. This fieldwork was a process in which I, as the researcher, became the student and learnt from the participants, not only about the research topic but also about how to behave and 'be' in the Solomon Islands context. The chapter concluded with comment on the procedures taken to analyse the data and ensure dependability of the findings which are presented in the following chapter.

I would like to close this chapter with the comment one participant made at the conclusion of our stori time together which reflects the power of the research

relationship to give participants space to speak out their dreams, insights and concerns. It also highlights the desire, of both those who participated in this study and myself, for this research to contribute to bringing about positive change for young people in the Solomon Islands.

"Just to say thank you to you for the work you are doing; because it kind of draws out in us what we have, as youth workers, in what we want to see happening in our country for our young people. So they [our thoughts] have been with us and we have been keeping them to ourselves, but when such opportunities come we share them and when we share them we feel that we are doing something that later on, in years to come, will actually bring forth results for our young people" Male Youth Leader, Honiara.

The following chapter details what was shared by participants, with the hope that it may play a part in 'bringing forth results for young people'.

THE THINNING, THICKENING AND ARTICULATION OF